African Americans experience rates of diabetes complications and hospitalization which are between 1.5 and 4 times greater than white patients. Glycemic control reduces risk of diabetes complications and hospitalization, and can be improved through medication as well as lifestyle changes. No trial of lifestyle has achieved sustained improvements in glycemic control in low-income African Americans. This project is a randomized controlled trial to test an innovative lifestyle intervention to achieve sustained improvements in glycemic control among low-income African American diabetes patients. The LIFE (Lifestyle Improvement through Food and Exercise) program is a diabetes self-management program focused on diet and exercise, informed by anthropological research on models of food and health among low-income African-Americans. Pilot work demonstrated that the LIFE Program is effective in improving glycemic control among low-income African Americans at 6-months. The main goal of the current study is to determine whether the LIFE Program can achieve sustained improvements in glycemic control for 12 months. The trial will randomize low-income African American adults with diabetes to a control group, which receives standard diabetes education, or an intervention group, which receives the LIFE Program, featuring a 6-month intervention (20 group meetings with peer support telephone calls) followed by an 18-month maintenance phase (monthly peer support phone calls and quarterly group sessions). The primary aim of the proposed research is to compare low-income African American diabetes patients receiving the LIFE Program with those in a standard of care control group on change in glycemic control at 12 months. Our primary hypothesis is that patients in the intervention group will achieve a change in A1c from baseline that is less than patients in the control group. Secondary aims are to compare low-income African American diabetes patients receiving the LIFE Program with those in a standard of care control group on (a) change in glycemic control at 24 months; (b) change in physical activity and total energy intake at 12 months; (c) change in physical activity and total energy intake at 24 months; and (d) to obtain estimates needed for a subsequent trial, including weight, blood pressure, and diabetes-related hospitalizations. For secondary aims we hypothesize that a) the intervention group will achieve a mean 24- month change in A1C that is less than the change in the control group; b) at 12 months, a greater proportion of intervention patients will have achieved the activity goal of 150 minutes of moderate activity per week, and the intervention group will achieve a greater reduction from baseline in mean total energy intake than the control group; and c) at 24 months, a greater proportion of intervention patients will have achieved the activity goal of 150 minutes of moderate activity per week, and the intervention group will achieve a greater reduction from baseline in mean total energy intake than the control group.